Abstract

Wind energy development in sandy, Gobi, and desertregions presents significant advantages; however, therelationship between wind turbines/wind farms and dusttransport remains unclear. This study pioneers the use ofscaled single and multiple wind turbine models in asimulated aeolian environment within an atmosphericboundary layer wind tunnel. By employing particles ofvarying sizes and concentrations to representsandstorms of different intensities, we investigate theimpact of wind turbines/farms on dust transport and theeffect of wind farm wakes on erodible surfaces. Theresults demonstrate that rotating turbine bladessignificantly obstruct dust transport, with tip vortices andthe central vortex causing localized dust accumulation—particularly pronounced in upper tip vortex regions.Furthermore, dust accumulation occurs above windfarms, with the accumulation area expanding as particlesize decreases. Operational wind farms promote dustdeposition, with deposition rates decreasing as particlesize increases. In the wake region, larger particlesaccumulate more readily, showing a positive correlationbetween accumulation and particle size. Additionally,large-scale wind farm wake effects effectively suppressdust emission.

Abstract

Solar power generation is a crucial pathway towardachieving global carbon neutrality. However, manycountries and regions lack a comprehensiveunderstanding of the barriers to solar developmentfrom a global perspective. To address this issue, weselected 11 representative countries (or regions)worldwide and projected the electricity consumptionand solar land demand of 665 first-level administrativeunits by 2050. Based on land, capital, and policydimensions, we constructed PV development constraintscenarios and analyzed the developmental constraintsand future pathways of solar power generation in theseregions. The results indicate that: (1) Global electricityconsumption and land demand will continue to grow.By 2050, only a few countries—such as Australia, Egypt,and South Africa—can largely meet their electricitydemand using barren land alone, while only 19.94% ofglobal first-level administrative units can do so; mostregions must rely on composite land-use approachessuch as construction land integration. (2) Based oncombinations of land availability and economicconditions, PV development constraints can becategorized into four types: land-constrained, capital-constrained, land and capital-constrained, and policy-constrained. (3) Japan, South Korea, the EuropeanUnion, Brazil, China, and the United States are land-constrained; Egypt and South Africa are capital-constrained; Thailand and India are land and capital-constrained; and Australia is policy-constrained. (4) Atthe first-level administrative unit scale, land-constrained areas are the most widespread (67.92%),while capital-constrained areas are the least common(4.05%). This classification framework provides ascientific theoretical basis and practical guidance forformulating differentiated and targeted regional PVdevelopment strategies.

Abstract

Virtual Inertia Damping Control (VID-Control) is aneffective method to improve the inertia and damping ofDC microgrids. However, unreasonable design ofadditional inertia (J) and damping (D)parameters willlead to severe system oscillations or even instability. Toaddress this issue, this paper proposes a VID-Controlstrategy for multi-bus DC microgrids based on anextended observer. Leveraging Lyapunov stabilitytheory, this strategy enables the decentralized design ofvirtual inertia, damping and the low-level fast voltagecontrol. It does not require measurements of theconverter’s own output current or parameters of otherconverters. This not only realizes decentralized controldesign of VID-Control in multi-bus DC microgrids but alsoreduces the cost of output current sensors whileenhancing the system’s inertia and damping. Theeffectiveness of the proposed method is verified throughsimulations of multi-DC buck converter system.

Abstract

The operation of manufacturing plants (MPs) resultsin considerable energy consumption and carbonemissions. Existing studies typically adopt static carbonemission factors when estimating the carbon footprintsof MPs, overlooking their spatiotemporal variabilityacross different nodes of the power distribution network(DN). Thus, in this paper, we propose a two-layer energymanagement framework that explicitly accounts fordynamic carbon emission factors to achieve coordinatedlow-carbon operation between the DN and MPs. In theproposed framework, the upper-level model formulatesan electricity-carbon sharing problem among MPs,aiming to achieve cooperative utilization of renewableenergy and carbon allowances while minimizing the totaloperation cost. The lower-level model addresses theoptimal power flow of the DN, where the carbonemission flow (CEF) theory is employed to calculate thenode-level carbon emission factors. These factors arethen integrated with locational marginal prices (LMPs) toconstruct an electricity-carbon integrated pricingmechanism, which dynamically guides the upper-levelenergy scheduling and sharing decisions of MPs. Casestudies demonstrate that the proposed two-layer energymanagement framework can effectively reduce bothoperation costs and carbon emissions for MPs.

Abstract

This paper presents the implementation and analysisof a ML-based MPC system integrated with a real-timeoccupant feedback mechanism through field tests in acommercial building. We demonstrate the efficacy of theoccupant-centric control strategy that dynamicallyadjusts thermal comfort setpoints based on direct userinput. Furthermore, we propose a comprehensiveframework for analyzing the resulting data, offeringinsights into system performance, energy consumption,and occupant satisfaction. The findings illustrate thatincorporating a human-in-the-loop approach canenhance building energy efficiency withoutcompromising occupant comfort.

Abstract

To address the influence of converter control andtransition resistance on traditional fault-locationmethods for renewable energy export lines, this paperproposes a single-terminal fault location method basedon active injection of complex-frequency signals.Leveraging the controllability of a Static Var Generator(SVG), dual characteristic-frequency signals are injectedimmediately after fault occurrence. Stable injection isachieved through frequency-separation control andconstant-current amplitude regulation. Equivalent faulttopologies at the characteristic frequencies areestablished for both phase-to-phase and single-phase-to-ground faults, from which complex-valued distanceequations are derived. Considering the long-line effects,a distributed-parameter model is further adopted toconstruct dual-frequency distance equations, and theSpider Wasp Optimizer (SWO) is introduced to achieveinitial-value-independent fault distance estimation.Simulation results in PSCAD/EMTDC demonstrate thatthe proposed method exhibits strong robustnessagainst transition resistance and is well-suited for high-accuracy fault location in renewable energy exportsystems.

Abstract

This study numerically investigates the multiphaseflow characteristics inside a hollow rotating shaft undernear-vacuum conditions, aiming to clarify thefundamental flow mechanisms governing rotor cooling inflywheel energy storage systems (FESS). Computationalfluid dynamics (CFD) simulations were conducted tocompare the stationary and rotating configurations,focusing on the influence of shaft rotation on phasedistribution, pressure evolution, and velocity fieldcharacteristics. In the stationary case, the impinging jetinduces a strong local pressure peak and downward flow,accompanied by pronounced air entrainment,incomplete gas removal, and localized cavitation neargeometric transitions. When rotation is introduced,centrifugal force dominates the flow dynamics, leadingto a stable annular flow structure with liquid adhering tothe wall and air confined to the shaft core. The air phasedevelops a helical backflow path driven by the low-pressure core region, while near-wall backflow andcavitation are significantly suppressed. The findingsreveal a transition from a gravity- and inertia-driven,chaotic two-phase regime to a rotation-dominated,stratified annular flow regime. This transition enhancesflow stability and liquid film continuity, offeringimportant insights into the coupled flow-phaseinteractions under vacuum and providing practicalguidance for the thermo-fluid optimization of high-speedflywheel rotors.

Abstract

Deep Q-network (DQN) has shown significantpotential in industrial control, yet its high explorationcosts and slow convergence during early deploymenthinder practical adoption. To address this, this studyproposes a data-driven pretraining framework for DQN,based on a multi-step temporal difference (TD)algorithm. Historical operation data and well-designedreward functions are employed to pretrain the agent,significantly reducing post-deployment exploration. Theproposed framework is demonstrated through a casestudy on boiler combustion control system. The agentbuilt on an enhanced recurrent neural network (RNN)architecture regulates coal feeding and air supply tomeet dynamic steam load demands. The effects ofdifferent reward formulations and TD steps onpretraining efficacy are analyzed. Results show thatagents pretrained with linear and nonlinear rewardsboth outperform historical control strategies. The 3-stepTD pretraining strategy outperforms single-step TD,achieving 83.4% consistency with expert actions andrealizing a 5.8% improvement in energy efficiencycompared to manual operation.

Abstract

Machine learning (ML)-based Model Predictive Control (MPC) has shown promise for smart buildings, with online ML model adaptation often integrated to enhance long-term performance. However, abrupt building thermodynamic changes, such as opening/closing windows, can hardly been captured by the conventional online ML model adaptation. An Adaptive Transfer-Learning-Assisted Modular Learning (ATML) framework combined with dataset truncation is proposed by this paper to address the challenge. A simulation case study using a residential building was conducted for evaluation. Results show that, compared with conventional online ML model adaptation, ATML reduces prediction errors by 32.4% after 2 hours of the systemic change and by 25.4% after 22 hours, thereby demonstrating its effectiveness in coping with abrupt systemic changes.

Abstract

In recent years, the Model Predictive Control (MPC) for building systems has received extensive attention, but few studies have applied machine learning (ML)-based MPC to split air conditioners. This paper presents a field implementation of an iterative ML-based MPC with a cloud-based framework to demonstrate its feasibility and energy-saving potential for split air conditioners. The measured results show that, compared with conventional PID control, MPC achieved an average energy-saving rate of 22.9%. In addition, iteratively updating the predictive model enables MPC to achieve enhanced thermal comfort. Hence, this paper demonstrates the energy-saving capability of ML-MPC for split air conditioners and potential scalable cloud-based deployment in residential buildings.